Metabolic & Cardiometabolic
metabolic/bariatric surgery causes type 2 diabetes remission
In plain terms: Does metabolic surgery cause type 2 diabetes remission?
Strong support Metabolic & Cardiometabolic
Part of: 🩺 bariatric / metabolic surgery
RefutedContestedStrong support
consensus score 0.83
Yes—RCTs and cohorts show surgery causes far higher T2D remission than medical therapy, but remission wanes over time and relapse is common.
📅 Last reviewed: 2026-07-15 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: All trials, pooled (Meta-analysis)
MechanismIn-vitroAnimalObservationalRCTMeta-analysis
How the studies fall
34 support 0 contradict 0 tested null 5 mixed · 39 sources, 32 independent groups
What the evidence shows
Metabolic surgery causes type 2 diabetes remission far more often than medical therapy, in roughly half to four-fifths of patients within one to five years, confirmed in randomised trials. Durability is the caveat: the advantage narrows, relapse is common, and by ten years only biliopancreatic diversion remained significantly ahead of medical care.
The evidence (44)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Tan WL, Loo GH, Muthkumaran G, Kosai NR 2026 · Obesity surgery | observational | mixed | low | Retrospective cohort, n=1478 (Malaysia), formal remission could not be ascertained (no medication data); at 1yr, 80.9% of diabetics reached HbA1c<6.5% and 47.8%<5.7%, but 5-year subgroup (7.2% of cohort) showed selective retention/attrition |
| Ikramuddin S, Korner J, Lee WJ, Thomas AJ, Connett JE, Bantle JP, Leslie DB, Wang Q, Inabnet WB, Jeffery RW, Chong K, Chuang LM, Jensen MD, Vella A, Ahmed L, Belani K, Billington CJ 2018 · JAMA | RCT | mixed | high | Diabetes Surgery Study 5yr: RYGB improved composite targets but durability waned—triple-endpoint benefit not sustained at 5yr. |
| Cui B, He J, Yao H 2026 · Obesity surgery | meta-analysis | mixed | high | Network meta-analysis of 10 RCTs, n=1004, >=10-year follow-up, MBS vs medical therapy. Only BPD/DS achieved a statistically significant advantage over medical therapy for diabetes remission (OR=24.91); other procedures showed potential but |
| Galvan JAA, Loo GH, Kosai NR, Yeong CH, Chinna K, Nordin RB, Lee KK, Mustafa N 2026 · Journal of Multidisciplinary Healthcare | observational | supports | moderate | Retrospective single-centre cohort of 180 Malaysian adults with obesity and type 2 diabetes who had sleeve gastrectomy or gastric bypass, followed 5 years. Mean HbA1c fell from 6.37% to 5.62% (p<0.001), and 131 of the 157 patients still followed at 5 years were below 6.0%. But baseline HbA1c already reflected diabetes treated with intensive regimens including insulin, and the paper does not report whether any patient stopped their diabetes medication, so what it measures is glycaemic control rather than diabetes remission. |
| Lau ST, Loo GH, Kong SH, Mustafa N, Kosai NR 2026 · Obesity surgery | observational | supports | moderate | Retrospective cohort, n=236 (surgery 115, medical 121), Malaysia, up to 5yr follow-up. Overall diabetes remission 56.5% in bariatric group; remission strongly predicted by disease severity (mild-moderate IMS 77.5% vs severe 8.6%), not proce |
| Ochs V et al 2026 · BJS open | observational | supports | moderate | Multicenter retrospective cohort, n=9652 (RYGB/SG, 11 European centres), external validation cohort from Sweden. ML classifier predicted T2D remission up to 5yr with high discrimination (F1 0.88-0.89); underlying cohort data reflect real re |
| Nor Hanipah Z, Boyer LM, Queen KE, Staiano AE, Drews KL, Albaugh VL, Katzmarzyk PT, Schauer PR 2026 · Obesity Surgery | observational | supports | low | Single-centre retrospective series of 76 patients aged 10-25 (mean age 22.6, mean BMI 49.4) operated on in Baton Rouge, Louisiana between 2020 and 2025; 32% of them, about 24 people, had type 2 diabetes beforehand. Among those with follow-up data, diabetes remission occurred in 94% (15/16), so roughly two thirds of the diabetic patients contributed a result, over a mean follow-up of 16 months (range 1-48). There was no comparison group, and the paper does not say what criteria defined remission. |
| S Abdelrahim H, Elnabil-Mortada A, Boshra Gerges W, M Elmaleh H 2026 · Obesity Surgery | observational | supports | moderate | Retrospective cohort, n=115, revisional OAGB after failed restrictive procedures, 5-year follow-up. T2DM remission rate 70.8% at 5 years. |
| d10-3389-fendo-2026-1849157 2026 · Frontiers in Endocrinology | observational | supports | low | Narrative review of metabolic surgery techniques for T2DM, summarizing comparative efficacy, predictors of outcome, and long-term postoperative effectiveness; not a systematic review/meta-analysis by explicit methodology. |
| Pedarnig L, Jedamzik J, Nixdorf L, Richwien P, Mairinger M, Bichler C, Langer FB, Gensthaler L, Krebs M, Prager G, Felsenreich DM 2026 · Obesity surgery | observational | supports | moderate | Retrospective cohort, n=111 patients aged >=65, Vienna, 5yr follow-up. Prevalence of type 2 diabetes decreased from 54.1% to 22.9% after MBS. |
| Moioli D, Cesana G, Ciccarese F, Uccelli M, Zanoni AAG, Giorgi R, Oldani A, Bonaldi M, Rubicondo C, Delcarro A, Lee YH, Olmi S 2026 · Journal of laparoendoscopic & advanced surgical techniques. Part A | observational | supports | low | Retrospective single-center cohort, n=30 (SASI as redo/revisional surgery after failed primary procedures), 12-month follow-up. T2DM remission rates 'markedly improved' with near-complete resolution of comorbidities at 12 months (no exact % |
| Afzal A, Pandey JR, Ashraf T, Bagri AJA, Kumari M, Singla S, Singla B, Anika NN, Irshad A 2025 · Cureus | meta-analysis | supports | moderate | Systematic review of 10 RCTs found bariatric surgery consistently outperforms medical therapy for durable diabetes remission and glycemic control. |
| Kirwan JP, Courcoulas AP, Cummings DE, Goldfine AB, Kashyap SR, Simonson DC, Arterburn DE, Gourash WF, Vernon AH, Jakicic JM, Patti ME, Wolski K, Schauer PR 2022 · Diabetes Care | RCT | supports | high | ARMMS-T2D consortium design/early data: metabolic surgery durably superior to medical/lifestyle for remission. |
| Zhen XM, Macdonald A, Woodham B, Ryan B, Hng TM, Nettel-Aguirre A, Azar E, Hanna M, Yap J, Paulose J, Li G, Markovic S, Twigg SM, Ahlenstiel G, Wong J 2026 · Diabetes research and clinical practice | observational | mixed | moderate | Cohort, n=80 (44 young-onset, 36 older-onset T2DM), Australia, 12-month follow-up after metabolic surgery. Complete remission markedly lower in young-onset T2DM (38.1%) vs older-onset (75.0%, p=0.002), though difference not significant afte |
| Jaworski P, Dowgiałło-Gornowicz N, Orłowski M, Franczak P, Karpińska I, Major P, Tarnowski W 2026 · Wideochirurgia i inne techniki maloinwazyjne = Videosurgery and other miniinvasive techniques | observational | supports | moderate | Multicenter retrospective cohort (BARI-10-POL), n=113, Poland, 10-year follow-up (median 10.9yr). T2DM remission in 70.8%, improvement in 16.8%, no change in 12.4%; longer diabetes duration predicted lower remission. |
| Velji-Ibrahim J, Radadiya D, Patel H, Hassan C, Pugliese N, Devani K, Sharma P 2026 · Endoscopy international open | meta-analysis | supports | moderate | Network meta-analysis of 15 RCTs, n=1548 adults, >=52 weeks follow-up, comparing bariatric surgery (RYGB, LSG, GCP, LAGB) and bariatric endoscopy. RYGB and LSG (surgical arms) achieved significant T2DM remission alongside ESG. |
| d10-1093-bjs-znag063-123 2026 · British journal of surgery | meta-analysis | supports | moderate | Systematic review (PRISMA) of 33 studies (RCTs and cohorts), PubMed/Embase/Cochrane 2005-2024. Remission rates 60-80% for RYGB and 50-70% for sleeve gastrectomy vs lower for gastric banding; earlier disease stage and younger age associated |
| Schauer PR, Bhatt DL, Kirwan JP, Wolski K, Aminian A, Brethauer SA, Navaneethan SD, Singh RP, Pothier CE, Nissen SE, Kashyap SR, STAMPEDE Investigators 2017 · N Engl J Med | RCT | supports | moderate | In STAMPEDE, 150 people with type 2 diabetes and a BMI of 27 to 43 were randomly assigned to intensive medical therapy alone or the same therapy plus gastric bypass or sleeve gastrectomy, and 134 were reassessed at 5 years. Diabetes remission - HbA1c 6.0% or less on no diabetes medication - was reached by 11 of 49 after bypass (22%) and 7 of 47 after sleeve gastrectomy (15%), versus 0 of 38 on medical therapy alone. The looser target of HbA1c 6.0% or less with or without medication was met by 29% and 23% versus 5%, but lost significance once all 150 randomized patients counted (P=0.08, 0.17). |
| Martínez-Montoro JI, Pinazo-Bandera JM, García-Serrano S, García-Almeida JM, Ocaña-Wilhelmi L, Moreno-Ruiz FJ, Soler-Humanes R, Ruiz-Campos N, Sancho-Marín R, García-Cortés M, Muñoz-Garach A, Fernández-García JC, García-Fuentes E, Tinahones FJ, Garrido-Sánchez L 2026 · Surgery for obesity and related diseases : official journal of the American Society for Bariatric Surgery | observational | supports | moderate | Retrospective two-center cohort, n=145 adults with T2D undergoing bariatric surgery, 1-year follow-up. Lower liver fibrosis risk score (HFS<.12) associated with 73.6% diabetes remission vs 52.1% in higher-risk group (RR=1.41). |
| Boutros T, Ghusn W, Klim J, El Ghazal N, Acra A, Laplante SJ, Ghanem OM 2026 · Obesity surgery | observational | mixed | moderate | Retrospective cohort, n=1323 with initial ORMC remission, primary MBS 2008-2022, >=36mo follow-up. T2D recurrence occurred in 16.2% of patients who had initially remitted; recurrence driven more by baseline severity/insulin dependence than |
| Sanchez-Cordero S, Lopez-Gonzalez R, Hermoza R, Pujol-Gebellí J 2026 · Surgery for obesity and related diseases : official journal of the American Society for Bariatric Surgery | meta-analysis | supports | moderate | Systematic review (PRISMA) of 7 studies, n=3735 with 5-year follow-up, OAGB as primary procedure. Pooled 5-year T2D remission rate 77.4%; authors note evidence remains limited/heterogeneous. |
| Mingrone G, Panunzi S, De Gaetano A, Guidone C, Iaconelli A, Capristo E, Chamseddine G, Bornstein SR, Rubino F 2021 · Lancet | RCT | supports | high | Ten-year follow-up of a randomised trial in Rome: 60 people with type 2 diabetes of more than five years' duration and a body-mass index of 35 or more were assigned to medical therapy, Roux-en-Y gastric bypass or biliopancreatic diversion, and 57 of the 60 were still followed a decade later. Of the 40 surgically treated, 15 (37.5%) stayed in remission throughout. Ten-year remission rates were 5.5% on medical therapy, 25.0% after gastric bypass and 50.0% after biliopancreatic diversion (p=0.0082). Of 34 in remission at two years, 20 (58.8%) relapsed, but still averaged HbA1c 6.7%. |
| Courcoulas AP, Patti ME, Hu B, Arterburn DE, Simonson DC, Gourash WF, Jakicic JM, Vernon AH, Beck GJ, Schauer PR, Kashyap SR, Aminian A, Cummings DE, Kirwan JP 2024 · JAMA | RCT | supports | high | Pooled analysis of 4 US randomized trials (ARMMS-T2D; 262 of 305 eligible participants, median 11 years' follow-up). At 7 years diabetes remission was 18.2% after bariatric surgery vs 6.2% with medical/lifestyle management (P=.02); at 12 years it was 12.7% vs 0.0% (P<.001). HbA1c fell 1.6% from baseline in the surgery group and 0.2% in the medical group, a between-group difference of -1.4% at 7 years and -1.1% at 12 years. Remission rates declined in both groups over time, but the surgical advantage persisted; 25% of the medical group later underwent surgery. |
| Teshaev O, Murodov A, Jumaev N, Rakhmonova G, Lim I 2026 · Obesity surgery | observational | supports | moderate | Retrospective cohort, n=127 (OAGB), 10-year follow-up with 100% retention. Complete T2DM remission (ADA criteria) achieved in 94.5% (120/127); all insulin-dependent patients discontinued insulin. |
| Abou-Sleiman R, Chan D, Chan K, Dodd B, Hopkins G, Finch R, Mitchell D 2026 · ANZ journal of surgery | observational | supports | moderate | Retrospective cohort, n=226, public hospital Queensland, 2-year follow-up (96.9% retained). Complete diabetes remission achieved in 54.8% of patients at 2 years; 65% required insulin preoperatively. |
| Celik Esmer A, Sever S, Kundakci B 2026 · Obes Surg | meta-analysis | supports | high | Network meta-analysis of 11 RCTs found all bariatric procedures significantly increased T2D remission versus medical/lifestyle therapy. |
| Kopanos S, Hoyer A, Stoye FV, Feldkamp J, Pouwels S, Pape-Köhler C 2026 · Surgery for obesity and related diseases : official journal of the American Society for Bariatric Surgery | meta-analysis | supports | high | Systematic review and meta-analysis, 29 studies, n=71,904 with MAFLD undergoing bariatric surgery. Pooled T2D remission rate 59%, alongside 70% steatohepatitis and 57% fibrosis remission. |
| Alum EU, Emeruwa AP, Obasi DC, Okoroh PN, Aniokete UC, Akwari AA, Uzor S, Eji MO, Amadi EB 2026 · Clinica chimica acta; international journal of clinical chemistry | observational | supports | low | Narrative review of gut microbiome-derived metabolites as predictors of bariatric surgery outcomes including T2DM remission; higher beneficial metabolite levels associated with greater remission likelihood, though evidence inconsistent acro |
| Takano H, Ebihara Y, Oe Y, Wada H, Nitta T, Nakamura A, Shichinohe T, Hirano S 2026 · Biomedical reports | observational | supports | low | Retrospective cohort, n=90 (LSG), Japan, 6-month follow-up. T2D remission/improvement rate was 75%; higher skeletal muscle mass associated with remission. |
| Ikramuddin S, Korner J, Lee WJ, Bantle JP, Thomas AJ, Connett JE, Leslie DB, Inabnet WB, Wang Q, Jeffery RW, Chong K, Chuang LM, Jensen MD, Vella A, Ahmed L, Belani K, Olofson AE, Bainbridge HA, Billington CJ 2016 · Diabetes Care | RCT | supports | moderate | Adding Roux-en-Y gastric bypass to lifestyle-medical management achieved treatment goals more often at 3 years in uncontrolled T2D. |
| Tan SYT, Liang TR, Chua JKL, Tan HC, Ho ETL, Kovalik JP, Goh OQM, Lim CH, Eng AKH, Chan WH, Lim EKW, Tan JTH, Foo AX, Lee PC 2026 · J Clin Med | observational | supports | low | Retrospective single-centre cohort of 184 adults in Singapore with a BMI of 50 or above who had sleeve gastrectomy or gastric bypass; 61 had type 2 diabetes at baseline, with a mean HbA1c of 8.0%. Among those still attending, diabetes remission (HbA1c under 6.5% off medication for 3 months) was 69.2% at 12 months (27 of 39 followed) and 76.7% at 24 months (23 of 30), and mean HbA1c fell to 5.5% then 5.7%. Remission did not differ significantly between the two operations. There was no comparison group and no adjusted analysis. |
| Susmallian S, Babis I, Raziel A 2026 · Diabetes Metab Res Rev | observational | supports | moderate | Retrospective single-centre cohort of 60 Israeli patients with a BMI of 60 or more, followed a mean of 72 months after bariatric surgery. The paper counts remission and improvement together: type 2 diabetes 77.8%, sleep apnoea 85.7%, fatty liver 73.9%, high cholesterol 70.0%, while blood pressure improved in only 29.1%. Because remission and improvement are pooled into one figure, the abstract does not show how many of the patients with diabetes actually reached remission, and two patients died (3.3%), one of them after revisional surgery. |
| Leyaro B, Howie L, Ali A, Carragher R 2026 · Obesity facts | observational | supports | moderate | Retrospective cohort, n=186 (SG/RYGB), UK. Remission rates 57.5% at 2yr and 36.4% at 5yr; of those remitted at 2yr, 33.4% relapsed by 5yr. |
| Mendes C, Carvalho M, Gregorio J, Palmeira A 2026 · Obes Surg | observational | supports | low | Retrospective single-centre cohort of 81 Portuguese women five years after Roux-en-Y gastric bypass. Twelve of the women with diabetes (66.7%) moved to a better disease-control category by five years, and mean fasting glucose stayed below baseline (94.1 to 92.2 mg/dL). But the paper reports no diabetes remission rate: it records only one patient, across diabetes and sleep apnoea together, who ended up living without disease, alongside a rise in patients controlled without medication. Diabetes control was unrelated to weight regain (p=0.957). |
| Si Y, Tu Y, Wang Y, Zhang H, Han X, Liu W, Zhang P, Ma X, Yu H, Bao Y 2026 · Obesity surgery | observational | supports | moderate | Retrospective cohort, n=75, 1-year follow-up. Early reduction in serum clusterin independently predicted T2D remission (defined as HbA1c<5.7% + fasting glucose<5.6mmol/L off medication), partially mediated by improved insulin sensitivity. |
| Sjöström L, Peltonen M, Jacobson P, Ahlin S, Andersson-Assarsson J, Anveden Å, Bouchard C, Carlsson B, Karason K, Lönroth H, Näslund I, Sjöström E, Taube M, Wedel H, Svensson PA, Sjöholm K, Carlsson LM 2014 · JAMA | observational | supports | high | In the Swedish Obese Subjects cohort, 343 surgery and 260 usual-care patients who already had type 2 diabetes were followed for a median 10 years. Two years on, 72.3% of surgical patients were in remission (219 of 303) versus 16.4% of controls (34 of 207). By 15 years remission had fallen to 30.4% (35 of 115) versus 6.5% (4 of 62) - still a wide gap, but most surgical remissions had relapsed. Only 47% of the surgery group and 41% of controls attended the 15-year examination, so those later rates rest on small attending subsets, and treatment was chosen rather than randomised. |
| Olsson LM, Borgeraas H, Chakaroun RM, Hofsø D, Hertel JK, Dwibedi C, Mitteregger M, Holst JJ, Tremaroli V, Hjelmesæth J, Bäckhed F 2026 · Nature metabolism | observational | supports | moderate | Subanalysis of the randomized Oseberg trial (RYGB n=39, SG n=38), 12-month microbiome profiling with 5-year T2D remission linkage. Surgery-induced microbiome changes correlated with GLP-1, beta-cell function, and 5-year T2D remission, indep |
| Czerwińska A, Karpińska I, Dowgiałło-Gornowicz N, Jaworski P, Major P, BARI-10-POL Collaborative Study Group 2026 · Sci Rep | observational | supports | moderate | BARI-10-POL, a Polish multicentre retrospective cohort: 89 patients whose starting BMI was 50 or more had bariatric surgery between 2007 and 2014 and were followed a mean of 11.1 years. Among those who had type 2 diabetes, 59.1% were in remission at that point, no different from patients who started below BMI 50; hypertension, sleep apnoea and reflux remitted in 61.5%, 57.1% and 75.0%. The abstract does not say how many of the 89 had diabetes to begin with, or how remission was defined, and everyone in the study had surgery, so there is no untreated comparison. |
| Lee Y, Lee HH, Seo HS, Lee CM, Son SY, Park YS, Kim SH 2026 · Annals of surgical treatment and research | observational | supports | moderate | Multicenter retrospective cohort, n=410 East Asian patients, 12-month follow-up. T2DM remission rates 47.6% (>=55yo) vs 80.5% (<55yo), both demonstrating meaningful metabolic improvement. |
| Choi SI, Heo Y, Ryu SW, Park S, Kwon Y, Ha TK, Lee YJ, Lee S, Seo KW, Kim SH, Park JY, Kim JH 2026 · Obesity surgery | observational | supports | moderate | Prospective multicenter cohort, n=116, Korea, up to 2-year follow-up. T2DM remission rates 80.7% at 1 year, increasing to 85.7% at 2 years. |
| Sjöholm K, Sjöström E, Carlsson LM, Peltonen M 2016 · Diabetes Care | observational | supports | moderate | SOS: gastric bypass improved glucose/insulin/HOMA-IR partly independent of weight change at 2 and 10yr. |
| Abdulrazzaq S, Ibrahim W, Al-Hashemi AA, Aladab A, Hassan SS, Hamamyh T, Elsherif MA, Abusabeib A, Al-Ahbabi T, Saliba M, El-Akhal L, Bashah M, Elhag W 2026 · Sleep & breathing = Schlaf & Atmung | observational | supports | moderate | Prospective cohort, n=48 (sleeve gastrectomy), 5-year follow-up. T2DM remission rate 71.4%; cardiometabolic and OSA benefits retained even in patients with weight regain (~31%). |
| Guo H, Wang G, Zhang J, Pang L, Gu J 2026 · Clinics (Sao Paulo, Brazil) | observational | supports | low | Retrospective cohort, n=60 adolescents (Grade IV/V obesity) undergoing LSG, 31 completed 1-year follow-up. All patients with preoperative T2DM (8/8) achieved remission criteria at 1 year. |
| Jiao Y, Dong S, Liu Z, Zhao X, Wang K 2026 · Scientific Reports | animal | supports | low | Rat model of obese T2DM (sleeve gastrectomy vs sham). SG significantly reduced fasting glucose and insulin resistance versus sham, via IL-27/p38 MAPK/PGC-1alpha pathway in adipose tissue. |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.